Vehicle motion control device, on-vehicle system, and vehicle motion control method

Abstract

Provided is a vehicle motion control device capable of realizing, with a low processing load, generation of a travel trajectory that enables obtainment of a sense of security and comfort in accordance with individual preferences. This host vehicle motion control device comprises: an information acquisition unit that acquires host vehicle information pertaining to a host vehicle and environmental information pertaining to a surrounding environment of the host vehicle; a trajectory generation parameter prediction unit that inputs the host vehicle information and the environmental information acquired by the information acquisition unit into a trained model, and outputs trajectory generation parameters including a range of physical quantities indicating a motion state of the host vehicle and a range of a travelable region of the host vehicle; and a trajectory generation unit that inputs the trajectory generation parameters and generates a trajectory along which the host vehicle travels.

Description

Vehicle motion control device, in-vehicle system, and vehicle motion control method

[0001] The present invention relates to a vehicle motion control device, an in-vehicle system, and a vehicle motion control method for controlling the motion of a host vehicle.

[0002] One type of vehicle motion control technology, typified by driver assistance and autonomous driving, is a technology that generates a driving trajectory consisting of information such as the vehicle's target driving route and driving speed, and controls the powertrain, brakes, steering, etc. so that the vehicle travels along that trajectory.

[0003] Furthermore, as a method for generating a driving trajectory, for example, Patent Document 1 discloses a method for selecting one driving trajectory from multiple driving trajectories generated using multiple models depending on the difficulty of recognizing the vehicle's surrounding environment.

[0004] Japanese Patent Application Laid-Open No. 2021-160532

[0005] However, the method of generating a running trajectory in Patent Document 1 has a problem in that it generates a plurality of running trajectories using a plurality of models, which results in a large computational load.

[0006] In view of this, the present invention aims to provide a vehicle motion control device that, in order to reduce the computational load, inputs vehicle information and environmental information related to the vehicle's surrounding environment into a prediction model that has been trained on manual driving, etc., predicts trajectory generation parameters including the range of physical quantities that indicate the vehicle's motion state and the range of the vehicle's drivable area, and generates a single driving trajectory based on the predicted trajectory.

[0007] The vehicle motion control device includes an information acquisition unit that acquires host vehicle information about the host vehicle and environmental information about the host vehicle's surrounding environment, a trajectory generation parameter prediction unit that inputs the host vehicle information and the environmental information acquired by the information acquisition unit into a learned model and outputs trajectory generation parameters including a range of physical quantities that indicate the motion state of the host vehicle and a range of a driving area of ​​the host vehicle, and a trajectory generation unit that inputs the trajectory generation parameters and generates a trajectory for the host vehicle to travel.

[0008] According to the present invention, a prediction model that has been trained on manual driving, etc., is used to predict trajectory generation parameters including the range of physical quantities that indicate the vehicle's motion state and the range of the vehicle's drivable area, and a single driving trajectory is generated based on these, thereby reducing the processing load when generating the vehicle's driving trajectory.

[0009] 1 is a functional block diagram of an in-vehicle system according to a first embodiment. FIG. 2 is a hardware configuration diagram of a vehicle motion control device according to a first embodiment. FIG. 3 is a functional block diagram of a driving trajectory planning unit according to a first embodiment. FIG. 4 is a flowchart showing an overview of processing by the driving trajectory planning unit according to a first embodiment. FIG. 5 is a processing example of the functional blocks of the driving trajectory planning unit according to a first embodiment. FIG. 6 is a processing example of the functional blocks of the driving trajectory planning unit according to a second embodiment. FIG. 7 is another processing example of the functional blocks of the driving trajectory planning unit according to a second embodiment. FIG. 8 is a functional block diagram of an operation management unit according to a third embodiment. FIG. 9 is a functional block diagram of an in-vehicle system and an external system according to a fourth embodiment. FIG. 10 is a functional block diagram of an HMI unit according to a fifth embodiment. FIG. 11 is a display example of a display according to a fifth embodiment.

[0010] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The following description and drawings are examples for explaining the present invention, and some omissions and simplifications have been made as appropriate for clarity of explanation. The present invention can be implemented in various other forms. Unless otherwise specified, each component may be singular or plural.

[0011] In order to facilitate understanding of the invention, the position, size, shape, range, etc. of each component shown in the drawings may not represent the actual position, size, shape, range, etc. Therefore, the present invention is not necessarily limited to the position, size, shape, range, etc. disclosed in the drawings.

[0012] First, a first embodiment of the present invention will be described with reference to FIGS. 1 to 5. FIG.

[0013] <Functional Block Diagram of In-Vehicle System 1> FIG. 1 is a functional block diagram of an in-vehicle system 1 according to the present embodiment. The in-vehicle system 1 is mounted on a host vehicle V1 and is a system for performing vehicle motion control, such as driving assistance and autonomous driving. As shown in FIG. 1 , the in-vehicle system 1 according to the present embodiment includes an external communication device 11, a Global Navigation Satellite System (GNSS) 12, a map information storage unit 13, sensors 14, an HMI (Human Machine Interface) unit 15, a vehicle motion control device 2, a powertrain system 3, a brake system 4, and a steering system 5. The vehicle motion control device 2 according to the present embodiment also includes a traffic management unit 21, a driving trajectory planning unit 22, and a driving control unit 23. Each component will be described below in order.

[0014] <Group of information sources of vehicle motion control device 2> The external vehicle communication device 11 performs vehicle-to-vehicle communication between the vehicle V1 and other vehicles, or road-to-vehicle communication between the vehicle V1 and a roadside device, via wireless communication, and transmits and receives information about the vehicle, the surrounding environment, etc., and cloud information.

[0015] The GNSS 12 receives radio waves transmitted from artificial satellites such as quasi-zenith satellites and GPS (Global Positioning System) satellites, and acquires information such as the position of the vehicle V1.

[0016] The map information storage unit 13 stores general road information used in navigation systems, road information including information about curves such as road width and road curvature, information about road surface conditions and traffic conditions, information about the vehicle and surrounding environment, which is information about the driving conditions of other vehicles, etc. The information about the vehicle and surrounding environment and cloud information are updated successively using information acquired through vehicle-to-vehicle communication and road-to-vehicle communication via the exterior communication device 11.

[0017] The sensor 14 is an external environment recognition sensor, such as an image sensor, millimeter-wave radar, or lidar, that detects information about the vehicle and the surrounding environment, and also detects information such as driver operation, vehicle speed, acceleration, jerk, angular velocity, and wheel steering angle. The information about the vehicle and the surrounding environment detected by the external environment recognition sensor in the sensor 14 includes, for example, information about various objects present around the host vehicle V1, such as obstacles, signs, lane boundaries, lane outer lines, buildings, pedestrians, bicycles, and other vehicles. An example of an environmental element is a sign indicating a "bicycle-only road." The sensor 14 also recognizes lane boundaries, lane outer lines, and the like, based on the difference in brightness between the white lines in image data captured by the image sensor and the road surface.

[0018] The HMI unit 15 displays on the display and provides audio guidance from the speaker the information required by the user from information received through user input operations such as selection of a driving mode and setting of a destination, information acquired by the external communication device 11, the GNSS 12, and the sensors 14, and information recorded in the map information storage unit 13. The HMI unit 15 also issues an alarm to alert the user.

[0019] 2 is a hardware configuration diagram of the vehicle motion control device 2. As shown here, the vehicle motion control device 2 is an ECU (Electronic Control Unit) that controls the vehicle overall, and includes hardware such as a calculation device 24 such as a CPU (Central Processing Unit), a main storage device 25 such as a semiconductor memory, an auxiliary storage device 26, and a communication device 27. The calculation device 24 executes a program loaded into the main storage device 25, thereby realizing various functions of the traffic management unit 21 and the like. Note that in this embodiment, for convenience of explanation, the traffic management unit 21, the driving trajectory planning unit 22, and the driving control unit 23 have separate configurations, but they do not necessarily have to have separate configurations. When these units are used in an actual vehicle, the various functions of these units may be realized by a higher-level controller.

[0020] The vehicle driving modes controlled by the vehicle motion control device 2 include, for example, comfort mode, economy mode, sports mode, minimum time mode for minimizing travel time, minimum distance mode for minimizing travel distance, etc. These driving modes may be arbitrarily set by the user, may be set in advance by the user, or may be set by the operation management unit 21 based on travel situation information. In this way, the vehicle motion control device 2 sets the speed, acceleration, and jerk of the host vehicle V1, the speed and distance between the host vehicle V1 and a preceding vehicle traveling ahead of the host vehicle, etc.

[0021] <Operation management unit 21> Based on information acquired by the external communication device 11, the GNSS 12, and the sensors 14, and map information recorded in the map information storage unit 13, the operation management unit 21 generates information on the position information of the vehicle V1, information on various objects present in the vicinity of the vehicle V1 (information on the position and speed of the vehicle and the surrounding environment, etc.), and information on the behavior of the vehicle such as longitudinal acceleration, longitudinal jerk, lateral acceleration, yaw rate, and lateral jerk.

[0022] Furthermore, the operation management unit 21 periodically transmits the generated position information of the vehicle V1, information on various objects, and information on vehicle behavior to other vehicles and roadside devices via the exterior communication device 11, and also transmits the information to the map information storage unit 13. The map information storage unit 13 successively updates the stored map information using the acquired position information of the vehicle V1, information on various objects, and information on vehicle behavior.

[0023] Furthermore, the operation management unit 21 sets information on a route from the current position of the vehicle to the destination based on the information on the position of the vehicle V1, information on various objects, information on the behavior of the vehicle, and information (e.g., driving mode and destination) acquired by the HMI unit 15. This route information is displayed on a map, which will be described later, and is used to determine the vehicle's driving trajectory. The information generated and set by the operation management unit 21 will be described below as driving situation information.

[0024] <Travel control unit 23> The travel control unit 23 sets a target driving force, a target braking force, a target steering angle, etc., and controls the powertrain system 3, the brake system 4, and the steering system 5 so that the vehicle travels along the travel trajectory output from the travel trajectory planning unit 22.

[0025] <Group of Controlled Objects of Cruise Control Unit 23 > The powertrain system 3 controls the driving force generated by the internal combustion engine, the electric motor, etc., based on the operation by the driver and the target driving force output from the cruise control unit 23 .

[0026] The brake system 4 controls the braking force generated by the brake calipers and the like based on the operation by the driver and a target braking force output from the cruise control unit 23 .

[0027] The steering system 5 controls the steering angle of the wheels based on the operation by the driver and a target steering angle output from the cruise control unit 23 .

[0028] 3, the traveling trajectory planning unit 22 includes an information acquisition unit 22a, a trajectory generation parameter prediction unit 22b, a trajectory generation unit 22c, and an information output unit 22d. Processing in each unit will be described below with reference to the flowchart in FIG.

[0029] <<Step S1>> First, in step S1, the information acquisition unit 22a acquires from the operation management unit 21 host vehicle information relating to the specifications and position of the host vehicle V1 and external environment information relating to the surrounding environment of the host vehicle V1.

[0030] <<Step S2>> Next, in step S2, the trajectory generation parameter prediction unit 22b receives as input the host vehicle information and external environment information acquired in step S1, and predicts trajectory generation parameters using the trained model M. The trajectory generation parameters predicted here are information including a range of physical quantities including at least one of the speed, acceleration, and jerk of the host vehicle V1, and a range of a driveable area of ​​the host vehicle V1. Furthermore, the range of physical quantities and the range of the driveable area of ​​the host vehicle V1 are predicted for at least one of the road shape, road surface conditions, and obstacles present around the host vehicle V1, which are included in the external environment information.

[0031] The trained model M is a model trained to output trajectory generation parameters using input data measured using the GNSS 12, the sensors 14, etc. when an occupant of the host vehicle V1 manually drives the host vehicle V1. This trained model M is sufficiently trained, for example, based on machine learning. In the case of supervised learning, the training data is, for example, a dataset in which the range of correct physical quantities of the host vehicle V1 and the range of the correct drivable area of ​​the host vehicle V1 that the trained model M should output are labeled with respect to the behavior of the host vehicle V1 when an occupant of the host vehicle V1 manually drives the host vehicle V1 and the surrounding environment of the host vehicle V1. Furthermore, the trained model M may be a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), a long short-term memory (LSTM), or a combination thereof.

[0032] <<Step S3>> In step S3, the trajectory generation unit 22c calculates the target route W and target speed Ve of the host vehicle V1 based on the host vehicle information and external environment information acquired in step S1 and the trajectory generation parameters predicted in step S2.

[0033] The trajectory generation unit 22c is, for example, a model that uses an optimization method to obtain an operation amount for approaching a target as close as possible to the target for the target problem. The trajectory generation unit 22c may also be a model that uses rule-based, Bayesian estimation, maximum likelihood estimation, or a combination of these.

[0034] <<Step S4>> In step S4, the information output unit 22d outputs information about the traveling trajectory of the host vehicle V1, which is composed of the target route W and the target speed Ve generated in step S3, to the traveling control unit 23. This allows the traveling control unit 23 to control the host vehicle V1 with an appropriate target route W and target speed Ve that take into account the relationship between moving objects and environmental factors around the host vehicle.

[0035] <Example of Processing by Travel Trajectory Planning Unit 22> Here, with reference to FIG. 5, the processing of steps S2 to S3 in a situation where a pedestrian P is present in the traveling direction of the host vehicle V1 will be described.

[0036] Figure 5 illustrates the target route W and target speed Ve within the drivable area Av of the vehicle V1 when the vehicle V1 is traveling on a straight road sandwiched between the outer roadway lines Lm1 and Lm2 and a pedestrian P is walking in the direction of arrow D near the outer roadway line Lm2 ahead of the vehicle.

[0037] Here, the drivable area Av is an area in which the host vehicle V1 can travel without deviating from the outer lane lines Lm1 and Lm2 when the center of gravity of the host vehicle V1 is traveling on the boundary line of the drivable area Av and without colliding with the pedestrian P. The size and shape of the drivable area Av are determined based on what the trained model M has learned, as well as on the ability of the host vehicle V1 to avoid collision with obstacles and comply with regulations; the range may be wider or narrower than that shown in FIG. 5 as long as these requirements are met.

[0038] In this environment, the trajectory generation unit 22c generates a target path W within the drivable area Av that moves away from the pedestrian P. The target speed Ve at this time is determined according to the speed learned by the learned model M. For example, if learning has been performed to maintain a constant speed when avoiding the pedestrian P, the trajectory generation parameter prediction unit 22b outputs a prediction that the speed is Lim1, and the trajectory generation unit 22c generates the target speed Ve1 based on the prediction.

[0039] Furthermore, if learning has been performed to set the speed in the vicinity of pedestrian P (specifically, from distance Tr1 to distance Tr2) when avoiding pedestrian P to speed Lim2, which is lower than speed Lim1, the trajectory generation parameter prediction unit 22b outputs a prediction that the speed from distance Tr1 to distance Tr2 will be Lim2, and the trajectory generation unit 22c generates the target speed Ve2 based on this prediction.

[0040] If parameters optimized only for specific scenes are used, only target speeds with speed limits of Lim1 or Lim2 can be generated. However, by generating a driving trajectory that matches human senses based on the driver's learning results from his or her normal driving, not only can the comfort of the occupants of the vehicle V1 be maintained in a variety of scenes, but also highly safe driving can be achieved that is less likely to cause a sense of danger to pedestrians P when a low target speed Ve2 is generated.

[0041] In addition, in an emergency, such as when avoiding a collision with a pedestrian that has jumped out into the road, the driver may suddenly brake or steer. However, if such obviously abnormal driving is also included in the learning data, the quality of the trained model M will deteriorate. Therefore, it is desirable to exclude such obviously abnormal driving from the learning data to prevent deterioration in the quality of the trained model M.

[0042] As described above, according to this embodiment, when both environmental elements such as the outer lane line and a moving object such as a pedestrian are detected, by using trajectory generation parameters such as speed, acceleration, and jerk output from the trained model M trained using measurement data during manual driving, it is possible to generate an appropriate driving trajectory that is tailored to the preferences of the vehicle's occupants while maintaining a low processing load.

[0043] Next, a travel path planning unit 22 according to a second embodiment of the present invention will be described with reference to Figures 6 and 7. Note that a duplicated description of points common to the first embodiment will be omitted.

[0044] <<First Example (FIG. 6)>> Fig. 6 illustrates processing in a situation where the host vehicle V1 is traveling around an S-shaped curve surrounded by outer roadway lines Lm1 and Lm2. Fig. 6 illustrates the target route W within the drivable area Av of the host vehicle V1 and the longitudinal acceleration obtained by time-differentiating the target speed Ve in a situation where deceleration is required to enter the curve.

[0045] In this environment, the trajectory generation unit 22c generates a target route W that passes through the travelable region Av. The position where the target route W passes through the travelable region Av is determined by the content learned by the trained model M, and may be not only the center of the travelable region Av shown in FIG. 6 but also a linear shape with a smaller curvature.

[0046] The target speed Ve at this time is determined according to the timing of deceleration start learned by the learned model M. For example, if learning has been performed in which deceleration starts frequently from a distance Tr3 close to the entrance of a curve, the trajectory generation parameter prediction unit 22b outputs a prediction as a behavior pattern in which the longitudinal acceleration at which deceleration starts from the distance Tr3 is Lg3, and the trajectory generation unit 22c generates the target speed based on this prediction.

[0047] Furthermore, if learning has been performed that indicates that deceleration begins more frequently from a distance Tr4 that is closer to the vehicle V1 than from distance Tr3, the trajectory generation parameter prediction unit 22b outputs a prediction as a behavior pattern in which the longitudinal acceleration at which deceleration begins from distance Tr4 is Lg4, and the trajectory generation unit 22c generates a target speed based on this prediction.

[0048] By generating such a driving trajectory, it is possible to generate a driving trajectory that provides a high sense of security and matches the preferences of the occupants of the vehicle.

[0049] <<Second Example (FIG. 7)>> Figure 7 illustrates an example of processing in a situation where the host vehicle V1 is traveling in the right lane of a two-lane road with an outer lane Lm1, a lane boundary line Lm3, and a lane center line Lm4, and a leading vehicle V2 is traveling ahead of the host vehicle V1. Figure 7 illustrates a target route within the drivable area Av of the host vehicle V1 in a situation where the host vehicle V1 changes lanes from the right lane to the left lane to overtake the leading vehicle V2.

[0050] Here, the drivable area Av is an area in which the vehicle V1 can travel without deviating outside the outer lane line Lm1 and the lane center line Lm4 and without colliding with the preceding vehicle V2 when the center of gravity of the vehicle V1 travels on the boundary line of the drivable area Av.

[0051] Under this environment, the trajectory generation unit 22c generates a target route for overtaking the preceding vehicle V2 within the drivable area Av. For example, when learning has been performed in which the target route frequently passes near the preceding vehicle V2, the trajectory generation parameter prediction unit 22b outputs a predicted behavior pattern of a turn that passes near the boundary line of the drivable area Av, and the trajectory generation unit 22c generates the target route W1 based on the prediction. Also, for example, when learning has been performed in which the target route frequently passes near a position away from the preceding vehicle V2, the trajectory generation parameter prediction unit 22b outputs a predicted behavior pattern of a turn that passes through a position away from the boundary line of the drivable area Av, and the trajectory generation unit 22c generates the target route W2 based on the prediction.

[0052] By generating such a driving trajectory, it is possible to generate a driving trajectory that provides a high sense of security and matches the preferences of the occupants of the vehicle.

[0053] Next, a traffic control unit 21 according to a third embodiment of the present invention will be described with reference to Fig. 8. Note that the overlapping description of points common to the first embodiment will be omitted.

[0054] 8, the operation management unit 21 of this embodiment includes an information acquisition unit 21 a, a vehicle control unit 21 b, and an information output unit 21 c. The operation management unit 21 selects a vehicle control method in the vehicle control unit 21 b based on the information acquired in the information acquisition unit 21 a, and outputs the selected information to the driving control unit 23 from the information output unit 21 c.

[0055] The information acquired by the information acquisition unit 21a includes, for example, the driving mode, which includes automatic driving control of the vehicle V1, switching control between automatic driving and manual driving of the vehicle V1, and assist control for manual driving by the occupant of the vehicle V1.

[0056] The vehicle control unit 21b selects appropriate vehicle control based on the information acquired by the information acquisition unit 21a. For example, if the driver selects automatic driving control using the HMI unit 15 and the situation of the vehicle V1 and the surrounding environment allows for automatic driving, automatic driving control is selected; otherwise, manual driving or manual driving assistance control is selected.

[0057] Thereafter, the driving control unit 23 controls the vehicle based on the vehicle control selected by the vehicle control unit 21b and the driving trajectory generated by the trajectory generation unit 22c.

[0058] In this way, the vehicle control unit 21b of the operation management unit 21 can realize safe and less uncomfortable vehicle control by selecting appropriate vehicle control according to the conditions of the vehicle V1 and the surrounding environment.

[0059] Next, an in-vehicle system 1 and an external system according to a fourth embodiment of the present invention will be described with reference to Fig. 9. Note that a duplicated description of points common to the first embodiment will be omitted.

[0060] 9, an external learning processing device 6, which is an external system, is added to the external communication device 11 of the in-vehicle system 1 shown in FIG.

[0061] The external learning processing device 6 is a server or cloud that communicates with the vehicle motion control device 2 via the extra-vehicle communication device 11. This external learning processing device 6 acquires information about the host vehicle V1 and the surrounding environment via the extra-vehicle communication device 11, and outputs information about the trained model M that has been trained based on that information to the vehicle motion control device 2. Furthermore, for areas that have not been acquired or learned by the in-vehicle system 1 with which it is communicating, the external learning processing device 6 outputs a trained model M that has been interpolated using information acquired and learned by the in-vehicle system 1 of another vehicle.

[0062] When interpolating the learned model M using information acquired from other vehicles, it is desirable to use driving information from other drivers whose driving preferences match those of the driver of the subject vehicle V1. This allows the interpolated learned model M to be made to match the driving preferences of the driver of the subject vehicle V1. In this case, to identify other drivers whose driving preferences match those of the driver of the subject vehicle V1, for example, it is sufficient to select, from among the drivers registered in the external learning processing device 6, other drivers whose driving characteristics, represented by the upper limits of physical quantities (e.g., speed, acceleration, jerk) observed when driving the vehicle, are similar to those of the driver of the subject vehicle.

[0063] In this way, the external learning processing device 6 reduces the required performance of the ECU that implements the in-vehicle system 1 by placing learning, which has a large processing load, outside the in-vehicle system 1, thereby reducing costs. In addition, by interpolating unlearned areas, the situations in which autonomous driving control is possible can be expanded.

[0064] Next, an HMI unit 15 according to a fifth embodiment of the present invention will be described with reference to Fig. 10. Note that the same points as in the first embodiment will not be described again.

[0065] 10, the HMI unit 15 has an information acquisition unit 15a, an information selection unit 15b, and an information output unit 15c. In the HMI unit 15, the occupant of the vehicle V1 selects information in the information selection unit 15b based on the information acquired in the information acquisition unit 15a, and the information is output from the information output unit 15c to the travel trajectory planning unit 22.

[0066] The information acquired by the information acquisition unit 15a and displayed on the display includes, for example, information regarding update contents based on the learning results of the trained model M, and it is desirable that the information be understandable to the occupant regarding changes in travel time to the destination, increases or decreases in acceleration, changes in driving position, etc. before and after the update.

[0067] The information selection unit 15b allows the selection of at least two learned models, one before the update and one after the update, and the occupant of the vehicle V1 decides whether or not to update the learned model M selected based on the information before and after the update.

[0068] 11 shows an example of the display in this embodiment. As shown here, the display shows the characteristics of the trained model M before and after the update, allowing the occupant to select a preferred model and use it for subsequent vehicle control.

[0069] In this way, the driving trajectory planning unit 22 generates a driving trajectory based on trajectory generation parameters predicted by the learned model M selected by the occupant of the vehicle V1 in the information selection section 15b of the HMI unit 15, thereby realizing vehicle control that provides a high level of comfort and security in line with the preferences of the occupant of the vehicle.

[0070] The present invention is not limited to the above-described embodiments, and various modifications and combinations of other configurations are possible without departing from the spirit of the present invention. Furthermore, the present invention is not limited to those having all of the configurations described in the above-described embodiments, and includes those in which some of the configurations are deleted, some of other configurations are added, and some of other configurations are replaced with other configurations.

[0071] REFERENCE SIGNS LIST 1 In-vehicle system 2 Vehicle motion control device 21 Operation management unit 21a Information acquisition unit 21b Vehicle control unit 21c Information output unit 22 Travel trajectory planning unit 22a Information acquisition unit 22b Trajectory generation parameter prediction unit 22c Trajectory generation unit 22d Information output unit 23 Travel control unit 24 Arithmetic unit 25 Main memory device 26 Auxiliary memory device 27 Communication device 3 Powertrain system 4 Brake system 5 Steering system 6 External learning processing device 11 Exterior communication device 12 GNSS 13 Map information storage unit 14 Sensor 15 HMI unit 15a Information acquisition unit 15b Information selection unit 15c Information output unit V1 Host vehicle V2 Leading vehicle A Ｖ Travel area W1, W2, W3 Target route of the vehicle Ve1, Ve2 Target speed of the vehicle Lg3, Lg4 Target longitudinal acceleration of the vehicle P Pedestrian D Direction of travel of the pedestrian

Claims

1. A vehicle motion control device comprising: an information acquisition unit that acquires host vehicle information about the host vehicle and environmental information about the host vehicle's surrounding environment; a trajectory generation parameter prediction unit that inputs the host vehicle information and the environmental information acquired by the information acquisition unit into a trained model and outputs trajectory generation parameters including a range of physical quantities that indicate the host vehicle's motion state and a range of the host vehicle's drivable area; and a trajectory generation unit that inputs the trajectory generation parameters and generates a trajectory for the host vehicle to travel.

2. A vehicle motion control device according to claim 1, wherein the trajectory generation unit is provided with a model based on at least one of rule-based, machine learning, optimization, Bayesian estimation, and maximum likelihood estimation.

3. A vehicle motion control device according to claim 1, characterized in that the trained model is a trained model generated by learning using the vehicle information and environmental information acquired during manual driving of the vehicle.

4. A vehicle motion control device according to claim 1, characterized in that the trajectory generation parameter prediction unit outputs the range of the physical quantities of the vehicle and the range of the drivable area for at least one of the road shape, road surface conditions, and obstacles acquired by the information acquisition unit.

5. A vehicle motion control device as described in claim 1, wherein the trajectory generation parameter prediction unit outputs a behavior pattern indicating the motion of the host vehicle relative to at least one of the road shape, road surface conditions, and obstacles acquired by the information acquisition unit, and the trajectory generation unit generates the trajectory using the behavior pattern as input.

6. A vehicle motion control device according to claim 5, wherein the behavior pattern represents at least one of acceleration, deceleration, and turning of the host vehicle.

7. A vehicle motion control device according to claim 1, wherein the physical quantity of the host vehicle includes at least one of the speed, acceleration, and jerk of the host vehicle.

8. A vehicle motion control device according to claim 1, characterized in that it comprises a vehicle control unit that performs at least one of the following based on the trajectory generated by the trajectory generation unit: automatic driving control of the host vehicle; switching control between automatic driving and manual driving of the host vehicle; and assist control of manual driving by the driver of the host vehicle.

9. A vehicle motion control device according to claim 1, characterized in that the information acquisition unit communicates with an external learning processing device that outputs the learned model obtained by interpolating unlearned areas of the learned model using information from manual driving by another driver.

10. A vehicle motion control device as described in claim 1, comprising an information selection unit that displays information regarding update contents based on the learning results of the trained model and allows an occupant of the vehicle to select whether or not to allow the trained model to be updated.

11. An in-vehicle system comprising the vehicle motion control device according to claim 1, a powertrain system that controls the driving force of the vehicle, a brake system that controls the braking force of the vehicle, and a steering system that controls the steering angle of the wheels of the vehicle, wherein the vehicle motion control device controls the powertrain system, the brake system, and the steering system so that the vehicle travels along the trajectory generated by the trajectory generation unit.

12. A vehicle motion control method comprising: an information acquisition step of acquiring host vehicle information and environmental information related to the host vehicle's surrounding environment; a trajectory generation parameter prediction step of inputting the acquired host vehicle information and environmental information into a learned model and outputting trajectory generation parameters including a range of physical quantities indicating the host vehicle's motion state and a range of the host vehicle's drivable area; and a trajectory generation step of inputting the trajectory generation parameters and generating a trajectory for the host vehicle to travel.

Images